Doc/lib/libcollections.tex

\section{\module{collections} ---
         High-performance container datatypes}

\declaremodule{standard}{collections}
\modulesynopsis{High-performance datatypes}
\moduleauthor{Raymond Hettinger}{python@rcn.com}
\sectionauthor{Raymond Hettinger}{python@rcn.com}
\versionadded{2.4}


This module implements high-performance container datatypes.  Currently,
there are two datatypes, deque and defaultdict.
Future additions may include balanced trees and ordered dictionaries.
\versionchanged[Added defaultdict]{2.5}

\subsection{\class{deque} objects \label{deque-objects}}

\begin{funcdesc}{deque}{\optional{iterable}}
  Returns a new deque object initialized left-to-right (using
  \method{append()}) with data from \var{iterable}.  If \var{iterable}
  is not specified, the new deque is empty.

  Deques are a generalization of stacks and queues (the name is pronounced
  ``deck'' and is short for ``double-ended queue'').  Deques support
  thread-safe, memory efficient appends and pops from either side of the deque
  with approximately the same \code{O(1)} performance in either direction.

  Though \class{list} objects support similar operations, they are optimized
  for fast fixed-length operations and incur \code{O(n)} memory movement costs
  for \samp{pop(0)} and \samp{insert(0, v)} operations which change both the
  size and position of the underlying data representation.
  \versionadded{2.4}
\end{funcdesc}

Deque objects support the following methods:

\begin{methoddesc}{append}{x}
   Add \var{x} to the right side of the deque.
\end{methoddesc}

\begin{methoddesc}{appendleft}{x}
   Add \var{x} to the left side of the deque.
\end{methoddesc}

\begin{methoddesc}{clear}{}
   Remove all elements from the deque leaving it with length 0.
\end{methoddesc}

\begin{methoddesc}{extend}{iterable}
   Extend the right side of the deque by appending elements from
   the iterable argument.
\end{methoddesc}

\begin{methoddesc}{extendleft}{iterable}
   Extend the left side of the deque by appending elements from
   \var{iterable}.  Note, the series of left appends results in
   reversing the order of elements in the iterable argument.
\end{methoddesc}

\begin{methoddesc}{pop}{}
   Remove and return an element from the right side of the deque.
   If no elements are present, raises an \exception{IndexError}.
\end{methoddesc}

\begin{methoddesc}{popleft}{}
   Remove and return an element from the left side of the deque.
   If no elements are present, raises an \exception{IndexError}.   
\end{methoddesc}

\begin{methoddesc}{remove}{value}
   Removed the first occurrence of \var{value}.  If not found,
   raises a \exception{ValueError}.
   \versionadded{2.5}
\end{methoddesc}

\begin{methoddesc}{rotate}{n}
   Rotate the deque \var{n} steps to the right.  If \var{n} is
   negative, rotate to the left.  Rotating one step to the right
   is equivalent to:  \samp{d.appendleft(d.pop())}. 
\end{methoddesc}

In addition to the above, deques support iteration, pickling, \samp{len(d)},
\samp{reversed(d)}, \samp{copy.copy(d)}, \samp{copy.deepcopy(d)},
membership testing with the \keyword{in} operator, and subscript references
such as \samp{d[-1]}.

Example:

\begin{verbatim}
>>> from collections import deque
>>> d = deque('ghi')                 # make a new deque with three items
>>> for elem in d:                   # iterate over the deque's elements
...     print elem.upper()	
G
H
I

>>> d.append('j')                    # add a new entry to the right side
>>> d.appendleft('f')                # add a new entry to the left side
>>> d                                # show the representation of the deque
deque(['f', 'g', 'h', 'i', 'j'])

>>> d.pop()                          # return and remove the rightmost item
'j'
>>> d.popleft()                      # return and remove the leftmost item
'f'
>>> list(d)                          # list the contents of the deque
['g', 'h', 'i']
>>> d[0]                             # peek at leftmost item
'g'
>>> d[-1]                            # peek at rightmost item
'i'

>>> list(reversed(d))                # list the contents of a deque in reverse
['i', 'h', 'g']
>>> 'h' in d                         # search the deque
True
>>> d.extend('jkl')                  # add multiple elements at once
>>> d
deque(['g', 'h', 'i', 'j', 'k', 'l'])
>>> d.rotate(1)                      # right rotation
>>> d
deque(['l', 'g', 'h', 'i', 'j', 'k'])
>>> d.rotate(-1)                     # left rotation
>>> d
deque(['g', 'h', 'i', 'j', 'k', 'l'])

>>> deque(reversed(d))               # make a new deque in reverse order
deque(['l', 'k', 'j', 'i', 'h', 'g'])
>>> d.clear()                        # empty the deque
>>> d.pop()                          # cannot pop from an empty deque
Traceback (most recent call last):
  File "<pyshell#6>", line 1, in -toplevel-
    d.pop()
IndexError: pop from an empty deque

>>> d.extendleft('abc')              # extendleft() reverses the input order
>>> d
deque(['c', 'b', 'a'])
\end{verbatim}

\subsubsection{Recipes \label{deque-recipes}}

This section shows various approaches to working with deques.

The \method{rotate()} method provides a way to implement \class{deque}
slicing and deletion.  For example, a pure python implementation of
\code{del d[n]} relies on the \method{rotate()} method to position
elements to be popped:
    
\begin{verbatim}
def delete_nth(d, n):
    d.rotate(-n)
    d.popleft()
    d.rotate(n)
\end{verbatim}

To implement \class{deque} slicing, use a similar approach applying
\method{rotate()} to bring a target element to the left side of the deque.
Remove old entries with \method{popleft()}, add new entries with
\method{extend()}, and then reverse the rotation.

With minor variations on that approach, it is easy to implement Forth style
stack manipulations such as \code{dup}, \code{drop}, \code{swap}, \code{over},
\code{pick}, \code{rot}, and \code{roll}.

A roundrobin task server can be built from a \class{deque} using
\method{popleft()} to select the current task and \method{append()}
to add it back to the tasklist if the input stream is not exhausted:

\begin{verbatim}
def roundrobin(*iterables):
    pending = deque(iter(i) for i in iterables)
    while pending:
        task = pending.popleft()
        try:
            yield task.next()
        except StopIteration:
            continue
        pending.append(task)

>>> for value in roundrobin('abc', 'd', 'efgh'):
...     print value

a
d
e
b
f
c
g
h

\end{verbatim}


Multi-pass data reduction algorithms can be succinctly expressed and
efficiently coded by extracting elements with multiple calls to
\method{popleft()}, applying the reduction function, and calling
\method{append()} to add the result back to the queue.

For example, building a balanced binary tree of nested lists entails
reducing two adjacent nodes into one by grouping them in a list:

\begin{verbatim}
def maketree(iterable):
    d = deque(iterable)
    while len(d) > 1:
        pair = [d.popleft(), d.popleft()]
        d.append(pair)
    return list(d)

>>> print maketree('abcdefgh')
[[[['a', 'b'], ['c', 'd']], [['e', 'f'], ['g', 'h']]]]

\end{verbatim}



\subsection{\class{defaultdict} objects \label{defaultdict-objects}}

\begin{funcdesc}{defaultdict}{\optional{default_factory\optional{, ...}}}
  Returns a new dictionary-like object.  \class{defaultdict} is a subclass
  of the builtin \class{dict} class.  It overrides one method and adds one
  writable instance variable.  The remaining functionality is the same as
  for the \class{dict} class and is not documented here.

  The first argument provides the initial value for the
  \member{default_factory} attribute; it defaults to \code{None}.
  All remaining arguments are treated the same as if they were
  passed to the \class{dict} constructor, including keyword arguments.

 \versionadded{2.5}
\end{funcdesc}

\class{defaultdict} objects support the following method in addition to
the standard \class{dict} operations:

\begin{methoddesc}{__missing__}{key}
  If the \member{default_factory} attribute is \code{None}, this raises
  an \exception{KeyError} exception with the \var{key} as argument.

  If \member{default_factory} is not \code{None}, it is called without
  arguments to provide a default value for the given \var{key}, this
  value is inserted in the dictionary for the \var{key}, and returned.

  If calling \member{default_factory} raises an exception this exception
  is propagated unchanged.

  This method is called by the \method{__getitem__} method of the
  \class{dict} class when the requested key is not found; whatever it
  returns or raises is then returned or raised by \method{__getitem__}.
\end{methoddesc}

\class{defaultdict} objects support the following instance variable:

\begin{datadesc}{default_factory}
  This attribute is used by the \method{__missing__} method; it is initialized
  from the first argument to the constructor, if present, or to \code{None}, 
  if absent.
\end{datadesc}


\subsubsection{\class{defaultdict} Examples \label{defaultdict-examples}}

Using \class{list} as the \member{default_factory}, it is easy to group
a sequence of key-value pairs into a dictionary of lists:

\begin{verbatim}
>>> s = [('yellow', 1), ('blue', 2), ('yellow', 3), ('blue', 4), ('red', 1)]
>>> d = defaultdict(list)
>>> for k, v in s:
        d[k].append(v)

>>> d.items()
[('blue', [2, 4]), ('red', [1]), ('yellow', [1, 3])]
\end{verbatim}

When each key is encountered for the first time, it is not already in the
mapping; so an entry is automatically created using the
\member{default_factory} function which returns an empty \class{list}.  The
\method{list.append()} operation then attaches the value to the new list.  When
keys are encountered again, the look-up proceeds normally (returning the list
for that key) and the \method{list.append()} operation adds another value to
the list. This technique is simpler and faster than an equivalent technique
using \method{dict.setdefault()}:

\begin{verbatim}
>>> d = {}
>>> for k, v in s:
	d.setdefault(k, []).append(v)

>>> d.items()
[('blue', [2, 4]), ('red', [1]), ('yellow', [1, 3])]
\end{verbatim}

Setting the \member{default_factory} to \class{int} makes the
\class{defaultdict} useful for counting (like a bag or multiset in other
languages):

\begin{verbatim}
>>> s = 'mississippi'
>>> d = defaultdict(int)
>>> for k in s:
        d[k] += 1

>>> d.items()
[('i', 4), ('p', 2), ('s', 4), ('m', 1)]
\end{verbatim}

When a letter is first encountered, it is missing from the mapping, so the
\member{default_factory} function calls \function{int()} to supply a default
count of zero.  The increment operation then builds up the count for each
letter.

The function \function{int()} which always returns zero is just a special
case of constant functions.  A faster and more flexible way to create
constant functions is to use \function{itertools.repeat()} which can supply
any constant value (not just zero):

\begin{verbatim}
>>> def constant_factory(value):
...     return itertools.repeat(value).next
>>> d = defaultdict(constant_factory('<missing>'))
>>> d.update(name='John', action='ran')
>>> '%(name)s %(action)s to %(object)s' % d
'John ran to <missing>'
\end{verbatim}

Setting the \member{default_factory} to \class{set} makes the
\class{defaultdict} useful for building a dictionary of sets:

\begin{verbatim}
>>> s = [('red', 1), ('blue', 2), ('red', 3), ('blue', 4), ('red', 1), ('blue', 4)]
>>> d = defaultdict(set)
>>> for k, v in s:
        d[k].add(v)

>>> d.items()
[('blue', set([2, 4])), ('red', set([1, 3]))]
\end{verbatim}